Add benchmarks for qmatmul operations (#2048)

* Add qmatmul bench

* add all dtypes

candle-core/benches/bench_main.rs

@@ -7,4 +7,5 @@ criterion_main!(
     benchmarks::random::benches,
     benchmarks::where_cond::benches,
     benchmarks::conv_transpose2d::benches,
+    benchmarks::qmatmul::benches,
 );